Cloning and expression of the dimethylsulfoniopropionate lyase gene dddY and the identification of the key amino acids necessary for its activity

Highlights

• •A new DMSP biodegradation strain was identified.
• •A new dimethylsulfoniopropionate lysis AsDddY was discovered and characterized.
• •Key amino acid residues of AsDddY were studied.
• •The diversity of arrangement of acu-dddY cluster was identified.

Abstract

The abundant solute dimethylsulfoniopropionate (DMSP) is crucial in marine ecosystems. In this study, a bacterium, Acinetobacter sp. ZS25, capable of completely mineralizing DMSP and producing DMS and acrylate, was isolated. The possible DMSP degradation pathway was primarily identified. The role of DMSP lyases DddY was identified through a combination of genomic comparison, gene knockout and heterologous expression. The K_m and k_cat of AsDddY for DMSP were 2.6 mM and 12.7 × 10³ s⁻¹, respectively. Site-directed mutagenesis was employed to examine the influence of specific amino acid residues (Thr131, Asp181, Tyr225, Gly230, Gly250, His263, His265, Glu269, Tyr271, Leu274, Tyr331, and His338) within AsDddY, elucidating their critical roles in the protein's functionality. Bioinformatics analysis revealed 19 distinct acu-dddY cluster order types, while the number of strains with a complete dddY-acu cluster is limited. Our findings offer novel insights into the mechanisms underlying DMSP biodegradation and enhance our understanding of the diversity of acu-dddY clusters present in natural bacterial populations.

Graphical abstract

1. Introduction

The sulfur-containing zwitterion dimethylsulfoniopropionate (DMSP) is produced in considerable quantities, estimated at 10⁹ tons per year globally, by a diverse array of marine phytoplankton species, including diatoms, dinoflagellates, and coccolithophores, in addition to certain macroalgal seaweeds and angiosperms. This compound plays a pivotal role in influencing various phenomena, ranging from cloud formation to avian behavior. (Andrew R J Curson et al., 2011; Ansede et al., 2001; A.R.J. Curson et al., 2011; Duyl et al., 1998). DMSP serves a dual function in both ecological and physiological contexts. It significantly contributes to nutrient cycling by providing a vital source of organic carbon and/or reduced sulfur for microbial communities (Dace et al., 1987; Dacey and Wakeham, 1986; Duyl et al., 1998; Yoch, 2002). Furthermore, DMSP performs several essential physiological roles, including functioning as an osmolyte, cryoprotectant, predator deterrent, antioxidant, chemoattractant, and as a protective agent against elevated hydrostatic pressure (Reisch et al., 2011; Stefels et al., 2007; Steudler and Peterson, 1984; Yancey, 2005). The catabolism of DMSP by microbial communities represents a major source of carbon and sulfur within marine ecosystems. Although a fraction of DMSP is degraded by algae (Sullivan et al., 2011; Sunda et al., 2002; Wolfe et al., 1997), it is predominantly hypothesized that bacteria are responsible for the majority of DMSP catabolism.

Extensive research has focused on microbial DMSP degradation, with studies by Curson et al. highlighting that marine bacteria primarily metabolize DMSP through various mechanisms (Andrew R J Curson et al., 2011; Wang and Gu, 2021; Yoch, 2002). To date, three metabolic pathways dominated by Proteobacteria and Firmicutes have been identified: demethylation, oxidation, and lysis (Wagner and Stadtman, 1962; Yoch, 2002). The nomenclature of these pathways is derived from the type of reaction catalyzed by the enzyme initiating the first step of DMSP catabolism. The demethylation pathway is initiated by the enzyme DMSP demethylase DmdA, leading to the degradation of DMSP into acetaldehyde and volatile methyl mercaptan (MeSH). This pathway constitutes the principal mechanism by which microorganisms metabolize DMSP, accounting for the catabolism of approximately 50 % to 90 % of DMSP (Reisch et al., 2008a; Zhang et al., 2019; Zheng et al., 2020). In the lysis pathway, DMSP is degraded by a variety of enzymes, including DMSP CoA-transferase DddD, DMSP-CoA ligase DddX, and eight distinct DMSP lyases (DddY, DddL, DddP, DddQ, DddW, DddK, DddU, and Alma1). While all these DMSP lyases possess the capability to cleave DMSP, they are characterized by unique protein sequences and are categorized into different superfamilies (Li et al., 2017; Thume et al., 2018; Wang et al., 2023a). DddD, with an estimated molecular weight of 93 kDa, is predominantly observed in gammaproteobacteria and operates as a class III CoA transferase. It facilitates the conversion of DMSP into 3-hydroxypropionyl-CoA (3-HP-CoA) and dimethyl sulfide (DMS). Conversely, DddX, identified in various Alphaproteobacteria, Gammaproteobacteria, and Firmicutes, belongs to the acyl-CoA synthase superfamily. DddX catalyzes the formation of the intermediate DMSP-CoA by reacting DMSP with coenzyme A, which is subsequently cleaved to produce DMS and acryloyl-CoA (Curson et al., 2008; Lei et al., 2018; Li et al., 2021). All the other DMSP lyases cleave DMSP producing acrylate and DMS (Fig. 1). In the oxidation pathway, DMSP is initially transformed into dimethylsulfoxoniopropionate (DMSOP). To date, no enzymes responsible for the oxidation of DMSP to DMSOP have been identified. However, researches have demonstrated that known DMSP lyases, DMSP CoA-transferase DddD and DMSP-CoA ligase DddX, can convert DMSOP into dimethylsulfoxide (DMSO) and acrylate, DMSO and 3-HP-CoA, and DMSO and acryloyl-CoA, respectively. It should be noted that acryloyl-CoA was not conclusively identifie (Carrión et al., 2023; Chhalodi and Dickschat, 2024; Chhalodiaa and Dickschat, 2023) (Fig. 1).

Fig. 1.

Biochemical pathways for DMSP degradation. Enzymes involved in the DMSP cleavage and demethylation pathways are shown. The demethylation of DMSP by DmdA produces MMPA, the oxidation of DMSP produces DMSOP, and in the lysis way, different lyases produce different products for breaking down DMSP, but DMS is produced in all. Dotted lines represent unconfirmed steps of the DddX DMSP cleavage pathway. THF, tetrahydrofolate; MMPA, methylmercaptopropionate; 3-HP, 3-hydroxypropionate; DMSOP, dimethylsulfoxonium propionate; DMSO, dimethylsulfoxide.

Among the eight DMSP lyases identified to date, Alma1 is the sole algal DMSP lyase, whereas the remaining seven have been isolated from bacterial source. Alma1 is characterized as a tetrameric enzyme that exhibits sensitivity to heavy oxygen and is classified within the aspartate racemase superfamily. However, its structural features and catalytic mechanism have yet to be fully elucidated (Johnston et al., 2016; Wang et al., 2023c). In contrast, DddP belongs to the M24 protease family and is associated with the Rosebacterium marine group (MRG) (Kirkwood et al., 2010). It is acknowledged as the most prevalent bacterial DMSP lyase within the marine metagenomic database. Conversely, DMSP lyases such as DddL, DddQ, DddW, DddY, DddU, and DddK belong to the cupin superfamily, which is distinguished by a highly functionally diverse proteome. Members of the cupin superfamily of DMSP lyases share two conserved copper-binding motifs and necessitate divalent metal ions as cofactors for their enzymatic activity (Lei et al., 2018; Todd et al., 2010). The majority of DMSP lyases in this superfamily are localized in the cytoplasm, with DddL being associated with the membrane and DddY situated in the periplasmic space (Andrew R J Curson et al., 2011).

The DddY enzyme is found within the phylum Proteobacteria, specifically in the β-, γ-, δ-, and ε-classes, and is distributed across both marine and terrestrial environments. Bacteria possessing the DddY enzyme are primarily situated in microaerobic sediments where DMSP is highly prevalent. Noteworthy genera include Arcobacter, Desulfovibrio, Shewanella, and Acinetobacter (Andrew R J Andrew R J Curson et al., 2011; Andrew R J Curson et al., 2011). DddY is recognized as the earliest identified DMSP lyase, first discovered in the betaproteobacterium Alcaligenes faecalis strain M3A. Notably, the characterization of the dddY gene was not accomplished until 2011, and it was only in 2017 that the structural and catalytic mechanisms of DddY were elucidated in the strain Acinetobacter bereziniae (Andrew R J Andrew R J Curson et al., 2011; Andrew R J Curson et al., 2011; Li et al., 2017). Nonetheless, there is a limited number of DMSP-degrading strains that possess acu-dddY clusters was reported, and research on acu-dddY pathways remains incomplete.

In this study, we isolated and characterized a DMSP-degrading bacterial strain, which they identified and designated as Acinetobacter sp. ZS25. Gas chromatography (GC) analysis revealed that ZS25 is capable of catabolizing DMSP and producing DMS. This paper further details the cloning and identification of the dddY gene, the assessment of the enzymatic properties of the DddY enzyme, and the identification of key amino acid residues. Additionally, the diversity of the dddY-acu gene cluster was investigated. This research provides a better understanding of the mechanism of DMSP biodegradation and the diversity of acu-dddY clusters contained in bacteria in nature.

2. Materials and methods

2.1. Chemicals, media and strains

DMSP and acrylate were procured from Solarbio, located in Beijing, China. All chemicals, solvents, and enzymes utilized in this study are commercially available. The processes of enrichment, isolation, purification, and cultivation of the isolates were conducted using Luria-Bertani broth (LB) or the basal medium (BM) as described by Yoch et al (Yoch, 2002) with 5 mM DMSP serving as the sole carbon source at 30°C. The strains and plasmids employed in this study are detailed in Table 1, while the sequences of the primers are provided in Table 2.

Table 1.

Strains and plasmids used in this study.

Strains and plasmids	Description	Source
Strains
Acinetobacter sp. ZS25	Strr, DMSP-degrading bacterium, Gram negative, wild type	This study
ZS25ΔdddY	Strr, dddY disruption mutant of ZS25	This study
ZS25ΔdddY/dddY	Strr Gmr, ZS25ΔdddY containing pBBR-dddY	This study
JQ135-dacuI	Strr Gmr, acuI disruption mutant of ZS25	This study
E. coli
DH5a	F2 recA1 endA1 thi-1 hsrdR17 supE44 relA1 deoRΔ(lacZYA-argF)U169 f 80lacZ ΔM15	Lab stock
BL21(DE3)	F2 ompT hsdS(rB2 mB2) gal dcm lacY1(DE3)	Lab stock
Plasmids
pBBR1MCS-5	Gmr, broad-host-range cloning plasmid	Lab stock
pET29a(+)	Kmr; expression plasmid	Lab stock
pJQ200SK	Gmr Mob+orip15A lacZa+ SacB, suicide plasmid	Lab stock
pJQΔdddY	Gmr, dddY gene deletion plasmid	This study
pJQ-dacuI	Gmr, acuI gene disruption plasmid	This study
pET-dddY	Kmr, fragment containing dddY gene inserted into pET29a(+)	This study
pET-dddY-P131A	Kmr; NdeI-XhoI fragment containing dddY-P131A inserted into pET29a(+)	This study
pET-dddY-D181A	Kmr; NdeI-XhoI fragment containing dddY-D181A inserted into pET29a(+)	This study
pET-dddY-Y225A	Kmr; NdeI-XhoI fragment containing dddY-Y225A inserted into pET29a(+)	This study
pET-dddY-G230A	Kmr; NdeI-XhoI fragment containing dddY-G230A inserted into pET29a(+)	This study
pET-dddY-G250A	Kmr; NdeI-XhoI fragment containing dddY-G250A inserted into pET29a(+)	This study
pET-dddY-H263A	Kmr; NdeI-XhoI fragment containing dddY-H263A inserted into pET29a(+)	This study
pET-dddY-H265A	Kmr; NdeI-XhoI fragment containing dddY-H265A inserted into pET29a(+)	This study
pET-dddY-E269A	Kmr; NdeI-XhoI fragment containing dddY-E269A inserted into pET29a(+)	This study
pET-dddY-Y271A	Kmr; NdeI-XhoI fragment containing dddY-Y271A inserted into pET29a(+)	This study
pET-dddY-L274A	Kmr; NdeI-XhoI fragment containing dddY-L274A inserted into pET29a(+)	This study
pET-dddY-Y331A	Kmr; NdeI-XhoI fragment containing dddY-Y331A inserted into pET29a(+)	This study
pET-dddY-H338A	Kmr; NdeI-XhoI fragment containing dddY-H338A inserted into pET29a(+)	This study
pBBR-dddY	Gmr, pBBR1MCS-5 harboring dddY	This study

Table 2.

Primers used in this study.

Primer	Sequence (5′–3′)	Description
kodddY-UF	AGCTTGATATCGAATTCCTGCAGTAGTTCAATAATTAGCTACTTTTCAAGGAGGACCG	To construct plasmid pJQ-ΔdddY
kodddY-UR	AATGCATGAATTGAATTACGTGGCTCGGCTAGTAAGGCTCT
kodddY-DF	AGAGCCTTACTAGCCGAGCCACGTAATTCAATTCATGCATTCCATGCAC
kodddY-DR	AGGGAACAAAAGCTGGAGCTCACACCTGATCATTACTATCCGATGTTGT
dacuI-F	AGCTTGATATCGAATTCCTGCAGGGTGTTGTAATTGAATCAAAACATCCAGC	To construct plasmid pJQ-dacuI
dacuI-R	AGGGAACAAAAGCTGGAGCTCGATAAATGGAGCAACTGTTGCAGGAAAATCC
expdddYF	AAGAAGGAGATATACCATGATGAATAAAAGAATTTTAGGTATAGTTTTTGGAACTACGC	To construct plasmid pET-dddY
expdddYR	CAGTGGTGGTGGTGGTGGTGCGGTTTCCAGTCGTGCAAAT
ExpdddY-P131A-F1	AAGAAGGAGATATACCATGATGAATAAAAGAATTTTAGGTATAGTTTTTGGAACTACGC	To construct pET-dddY-P131A
ExpdddY-P131A-R1	GTTTTTGCACTTCAGCGCTTGGCGCGTAGAGTGGGAACGCTTTATAAT
ExpdddY-P131A-F2	ATTATAAAGCGTTCCCACTCTACGCGCCAAGCGCTGAAGTGCAAAAAC
ExpdddY-P131A-R2	CAGTGGTGGTGGTGGTGGTGCGGTTTCCAGTCGTGCAAAT
ExpdddY-D181A-F1	AAGAAGGAGATATACCATGATGAATAAAAGAATTTTAGGTATAGTTTTTGGAACTACGC	To construct pET-dddY-D181A
ExpdddY-D181A-R1	GATATCCTTCGTCCAGTGTGCCTTAGTCGCGGTACGTGAAGTAATTGC
ExpdddY-D181A-F2	CAATTACTTCACGTACCGCGACTAAGGCACACTGGACGAAGGATATC
ExpdddY-D181A-R2	CAGTGGTGGTGGTGGTGGTGCGGTTTCCAGTCGTGCAAAT
ExpdddY-Y225-F1	AAGAAGGAGATATACCATGATGAATAAAAGAATTTTAGGTATAGTTTTTGGAACTACGC	To construct pET-dddY-Y225A
ExpdddY-Y225A-R1	CCCCAAGGCCCCCGCACTTCCGCATAAAGATATCCACCACGGAAAC
ExpdddY-Y225A-F2	GTTTCCGTGGTGGATATCTTGCGGCGGAAGTTATGGGGCCTTGGGG
ExpdddY-Y225A-R2	CAGTGGTGGTGGTGGTGGTGCGGTTTCCAGTCGTGCAAAT
ExpdddY-G230A-F1	AAGAAGGAGATATACCATGATGAATAAAAGAATTTTAGGTATAGTTTTTGGAACTACGC	To construct pET-dddY-G230A
ExpdddY-G230A-R1	CCCCAAGGCGCCATAACTTCCGCATAAAGATATCCACCACGGAAAC
ExpdddY-G230A-F2	GTTTCCGTGGTGGATATCTTTATGCGGAAGTTATGGCGCCTTGGGG
ExpdddY-G230A-R2	CAGTGGTGGTGGTGGTGGTGCGGTTTCCAGTCGTGCAAAT
ExpdddY-G250A-F1	AAGAAGGAGATATACCATGATGAATAAAAGAATTTTAGGTATAGTTTTTGGAACTACGC	To construct pET-dddY-G250A
ExpdddY-G250A-R1	TGAATAATTGTACAGTCATCGCAATTTCGGCCCCCACTTTTTCA
ExpdddY-G250A-F2	TGAAAAAGTGGGGGCCGAAATTGCGATGACTGTACAATTATTCA
ExpdddY-G250A-R2	CAGTGGTGGTGGTGGTGGTGCGGTTTCCAGTCGTGCAAAT
ExpdddY-H263A-F1	AAGAAGGAGATATACCATGATGAATAAAAGAATTTTAGGTATAGTTTTTGGAACTACGC	To construct pET-dddY-H263A
ExpdddY-H263A-R1	ATAAATCTCTTGAGGGTGGTGATACGCGTAGGGATAAGAGGTATTGAA
ExpdddY-H263A-F2	TTCAATACCTCTTATCCCTACGCGTATCACCACCCTCAAGAGATTTAT
ExpdddY-H263A-R2	CAGTGGTGGTGGTGGTGGTGCGGTTTCCAGTCGTGCAAAT
ExpdddY-H265A-F1	AAGAAGGAGATATACCATGATGAATAAAAGAATTTTAGGTATAGTTTTTGGAACTACGC	To construct pET-dddY-H265A
ExpdddY-H265A-R1	ATAAATCTCTTGAGGGTGCGCATAATGGTAGGGATAAGAGGTATTGAA
ExpdddY-H265A-F2	TTCAATACCTCTTATCCCTACCATTATGCGCACCCTCAAGAGATTTAT
ExpdddY-H265A-R2	CAGTGGTGGTGGTGGTGGTGCGGTTTCCAGTCGTGCAAAT
ExpdddY-E269A-F1	AAGAAGGAGATATACCATGATGAATAAAAGAATTTTAGGTATAGTTTTTGGAACTACGC	To construct pET-dddY-E269A
ExpdddY-E269A-R1	ATAAATCGCTTGAGGGTGGTGATAATGGTAGGGATAAGAGGTATTGAA
ExpdddY-E269A-F2	TTCAATACCTCTTATCCCTACCATTATCACCACCCTCAAGCGATTTAT
ExpdddY-E269A-R2	CAGTGGTGGTGGTGGTGGTGCGGTTTCCAGTCGTGCAAAT
ExpdddY-Y271A-F1	AAGAAGGAGATATACCATGATGAATAAAAGAATTTTAGGTATAGTTTTTGGAACTACGC	To construct pET-dddY-Y271A
ExpdddY-Y271A-R1	CGCAATCTCTTGAGGGTGGTGATAATGGTAGGGATAAGAGGTATTGAA
ExpdddY-Y271A-F2	TTCAATACCTCTTATCCCTACCATTATCACCACCCTCAAGAGATTGCG
ExpdddY-Y271A-R2	CAGTGGTGGTGGTGGTGGTGCGGTTTCCAGTCGTGCAAAT
ExpdddY-L274A-F1	AAGAAGGAGATATACCATGATGAATAAAAGAATTTTAGGTATAGTTTTTGGAACTACGC	To construct pET-dddY-L274A
ExpdddY-L274A-R1	ATAAACTTATTTTGATCAATACACTGCGGTTTGGTCGCTGTCATATAAATC
ExpdddY-L274A-F2	GATTTATATGACAGCGACCAAACCGCAGTGTATTGATCAAAATAAGTTTAT
ExpdddY-L274A-R2	CAGTGGTGGTGGTGGTGGTGCGGTTTCCAGTCGTGCAAAT
ExpdddY-Y331A-F1	AAGAAGGAGATATACCATGATGAATAAAAGAATTTTAGGTATAGTTTTTGGAACTACGC	To construct pET-dddY-Y331A
ExpdddY-Y331A-R1	GCATGGAATGCATGAATTGAATTACGTTCAAACGCGGTGAG
ExpdddY-Y331A-F2	CTCACCGCGTTTGAACGTAATTCAATTCATGCATTCCATGC
ExpdddY-Y331A-R2	CAGTGGTGGTGGTGGTGGTGCGGTTTCCAGTCGTGCAAAT
ExpdddY-H338A-F1	AAGAAGGAGATATACCATGATGAATAAAAGAATTTTAGGTATAGTTTTTGGAACTACGC	To construct pET-dddY-H338A
ExpdddY-H338A-R1	GCATGGAATGCCGCAATTGAATTACGTTCAAAGTAGGTGAG
ExpdddY-H338A-F2	CTCACCTACTTTGAACGTAATTCAATTCGCGCATTCCATGC
ExpdddY-H338A-R2	CAGTGGTGGTGGTGGTGGTGCGGTTTCCAGTCGTGCAAAT
qRT-dddY-F	ACTATAAAAAATTGGGCGGA	qRT-PCR
qRT-dddY-R	GCGGTTTGGTCAGTGTCAT
qRT-acuI-F	ATGCATTGGCTAAAGTATGAC
qRT-acuI-R	GACTGTTTCAGGTGATTGGCTT
qRT-dddC-F	GCTGAGCTGTAATAACAGGCCC
qRT-dddC-R	GTTGCAGTTGATACCTTGCTTACT
qRT-dddA-F	GCGATTTCATCGGGATCATCAC
qRT-dddA-R	CTCATGCCATAGAAGCTTATGTC
qRT-acuK-F	GCAGGAGGGACGATACGAGC
qRT-acuK-R	GGGCTGACATCGAAGAGATG
qRT-acuN-F	ACTTTCACGATGCTCATTGAG
qRT-acuN-R	AGATGTGCTTGGCAATCCACAATT
dddY-F	TGGCGGCCGCTCTAGAACTAGTATGAATAAAAGAATTTTAGGTATAGTTTTTGGAACTACG	To construct plasmid pBBR-dddY
dddY-R	TCGAATTCCTGCAGCCCGGGTTACGGTTTCCAGTCGTGCAAATCA

2.2. Enrichment and isolation of DMSP-degrading bacterium

The sediment sample was collected from the Zhoukou Park in Zhoukou city, Henan province, China. Approximately 10 g of sediment was added to BM medium containing 5 mM DMSP, and incubated at 30 °C for 7 d Subsequently, every 5 days, 10 mL of the enriched solution was transferred to fresh BM with 5 mM DMSP for subculturing. Following five subculturing processes, the enrichment cultures are sequentially diluted and subsequently plated onto solid BM medium supplemented with 5 mM DMSP. The colonies of different morphologies were selected for functional verification via gas chromatography (GC) analysis and high-performance liquid chromatograph (HPLC). The strains with DMSP-degrading ability, named ZS25, were selected for the further studies.

The 16S rRNA gene sequence of strain ZS25 was amplified through polymerase chain reaction (PCR) utilizing a pair of universal primers, 27F and 1492R. The resulting amplified sequence was subsequently purified and ligated into the pMD19-T vector for sequencing. Homologous sequences of the 16S rRNA gene from strain ZS25 were identified using the EZBiocloud database (https://www.ezbiocloud.net/). Phylogenetic analysis was conducted using MEGA version 7.0 software, employing the neighbor-joining method, with bootstrap values determined from 1000 replicates.

2.3. Analytical methods

The dimethyl sulfide (DMS) generated from DMSP was quantified using GC. Cultivation of the cells was conducted in serum bottles containing BM supplemented with 5 mM DMSP as the exclusive carbon source. These serum bottles were sealed with crimped rubber stoppers and incubated for a duration of 2 hours at 30°C. Subsequently, the headspace gases were analyzed for DMS content using a Shimadzu GC-2014 gas chromatograph equipped with a flame photometric detector, following the methodology outlined by Zhang (Zhang et al., 2014). DMS production is quantified as nmol DMS min⁻¹ mg protein⁻¹.

For the analysis of acrylate, supernatants from cell suspensions were prepared for HPLC analysis. These supernatants were centrifuged for 2 min and subsequently diluted at a ratio of 1:5 with an HPLC buffer consisting of 2.5 % acetonitrile and 0.2 % phosphoric acid in double-distilled water. The samples were analyzed using an HPLC system (UltiMate 3000, Thermo Fisher Scientific) equipped with a C18 reverse-phase column (4.6 × 250 mm, 5 µm), operated at a flow rate of 0.8 mL/min. The column was maintained at 40°C, with an injection volume of 20 μL. Detection of acrylate occurred at a wavelength of 214 nm (Fig. S1).

2.4. Genome sequencing and annotation

DNA extraction was performed utilizing the Ezup cfDNA Extraction Kit (Sangon Biotech) in accordance with the manufacturer's protocol. The entire genome of strain ZS25 was sequenced using the Illumina MiSeq platform at Biozeron, located in Shanghai, China. Gene identification and annotation were conducted using resources such as the National Center for Biotechnology Information (NCBI, https://www.ncbi.nlm.nih.gov/) and the Rapid Annotations using Subsystems Technology (RAST, https://rast.nmpdr.org/) database. The homologous genes involved in DMSP degradation were analyzed through NCBI, RAST, and additional databases.

2.5. Expression and purification of the his-tagged dddY

The dddY gene was amplified from the p-MD19T-dddY plasmid utilizing specific primers and subsequently cloned into the pET-29a plasmid to construct pET-dddY. This recombinant plasmid was transformed into Escherichia coli BL21 (DE3) cells, which were then induced with isopropyl β-D-1-thiogalactopyranoside (IPTG) at 16°C for a duration of 10 hours, following the attainment of an optical density at 600 nm (OD₆₀₀) of 0.6 in LB medium supplemented with kanamycin. Cell disruption was achieved through ultrasonic processing, and the resultant lysate was subjected to centrifugation to eliminate cellular debris. The recombinant DddY proteins were subsequently purified via nickel-nitrilotriacetic acid (Ni²⁺-NTA) agarose chromatography, followed by fractionation through anion exchange chromatography using a Source 15Q column and size-exclusion chromatography on a Superdex G75 column (GE Healthcare). The purity and molecular weight of the isolated protein were assessed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE).

2.6. Gene deletion and complementation

The primers utilized in this experiment are detailed in Table 2. The deletion of the dddY gene in A. sp. ZS25 was achieved through a two-step homologous recombination approach employing the suicide plasmid pJQ200SK. Homologous recombination-directing sequences were amplified and subsequently cloned into SacI/PstI-digested pJQ200SK, yielding the plasmid designated as pJQ-ΔdddY. This plasmid, pJQ-ΔdddY, was introduced into strain ZS25 via triparental mating facilitated by E. coli HB101(pRK2013). The desired mutant, designated as ZS25ΔdddY, was selected on LB plates following a double-crossover event.

The plasmid pBBR-dddY was engineered to facilitate gene complementation. The dddY gene was amplified utilizing the primers dddY-F and dddY-R, followed by ligation into the XhoI/HindIII-digested broad-host-range vector pBBR1-MCS5, resulting in the formation of pBBR-dddY. This plasmid was subsequently introduced into the mutant strain ZS25ΔdddY to produce the complemented strain ZS25ΔdddY/dddY.

2.7. Construction, expression and purification of site-directed amino acid mutation dddY

Mutant DddY proteins incorporating site-directed amino acid substitutions (P131, D181, Y225, G230, G250, H263, H265, E269, Y271, L274, Y331, and H338) were engineered, expressed, and purified through overlap-PCR to generate target fragments, which were subsequently cloned into the pET-29a plasmid. The resultant recombinant plasmids (pET-dddY-P131A, pET-dddY-D181A, pET-dddY-Y225A, pET-dddY-G230A, pET-dddY-G250A, pET-dddY-H263A, pET-dddY-H265A, pET-dddY-E269A, pET-dddY-Y271A, pET-dddY-L274A, pET-dddY-Y331A, and pET-dddY-H338A) were verified by sequencing prior to overexpression, employing the same methodologies as those used for the wild-type DddY.

2.8. Enzyme assays

The enzymatic activity of DddY and a series of mutant DddY proteins were determined by determining the production of acrylate by HPLC as described by Wang et al (Wang et al., 2023b). For routine assays, combine either the DddY protein or its mutant variants to achieve a final concentration of 0.5 nM, along with DMSP at a final concentration of 5 mM in a reaction buffer composed of 100 mM Tris–HCl at pH 8.0, ensuring a total volume of 200 μl. Incubate the mixture for 10 min at 50°C. Subsequently, terminate the reaction by adding perchloric acid, and quantify the acrylate present in the reaction mixture using HPLC. To ascertain the optimal temperature for DddY and its mutant proteins activity, reaction mixtures were incubated at temperatures ranging from 10°C to 70°C with a 10°C interval for 10 min each. The optimal pH for DddY and mutant DddY proteins was assessed at pH levels of 4.0, 5.0, 6.0, 7.0, 8.0, 9.0 and 10.0 using a wide-area buffer solution–Britton-Robinson Buffer, which is a mixture of 0.04 M H₃BO₃, 0.04 M H₃PO₄ and 0.04 M CH₃COOH (Li et al., 2017). In the aforementioned system, no enzyme is introduced as a negative control across the temperature range of 10°C to 70°C at the optimal pH 8.0. Similarly, no enzyme is incorporated as a negative control across the pH range of 4.0 to 10.0 at the optimal temperature 50°C. For kinetic parameter determination, all measurements were conducted at the enzyme’s optimal pH and temperature, using 0.5 nM DddY and DMSP concentrations ranging from 1 to 20 mM. To measure initial reaction rates, reactions were initiated by adding the DddY to the DMSP-containing buffer, and samples were collected at specific time points: 0, 2, 4, 6, 8, and 10 min post-initiation. Each sample was immediately mixed with perchloric acid to terminate the reaction, and acrylate levels were quantified by HPLC as described above. The initial reaction rate for each DMSP concentration was calculated from the linear phase of the reaction, expressed as mM of acrylate generated per minute. These initial rates were then subjected to non-linear regression analysis to determine the kinetic parameters. The values of K_m and V_max were obtained by fitting of a nonlinear regression curve to the Michaelis-Menten equation using Origin 8.0. The k_cat value was calculated by the following formula:

$$k_{cat}=\frac{V_{max}}{\left[\mathrm{E}\right]}$$

where V_max is the Maximum rate of enzyme-catalyzed reaction, [E] is the total concentration of enzymes in the reaction system. All reactions were performed in triplicate.

2.9. Database survey of the dddY-acu gene cluster

AsDddY was selected for comparisons against the nonredundant protein sequences database using BLASTP (protein-protein BLAST) on the NCBI website. The expected (E) value inclusion threshold was 10. Strains containing a AsDddY homolog with coverage of 40 % and identity of 40 % were collected and further assessed for the presence of AcuNK, DddAC, and DddR. The dddY-acu clusters with different arrangements were analyzed and plotted.

2.10. Real-time qPCR analysis

Cells of Acinetobacter sp. ZS25 were cultured in LB medium until reaching an optical density at OD₆₀₀ of 0.8. The cells were then harvested via centrifugation at 4000 × g, washed twice with BM, and subsequently resuspended in BM. The cell suspension, adjusted to an OD₆₀₀ of 1.0, was transferred into a 50-mL flask containing 20 mL of BM supplemented with either 5.0 mM DMSP or glucose as the sole carbon source. This setup was incubated at 180 rpm and 30°C. Following a 20-minute induction period, total RNA was extracted and complementary DNA (cDNA) was synthesized. Quantitative polymerase chain reaction (qPCR) was conducted using an Applied Biosystems real-time PCR system (Applied Biosystems, CA, USA) in conjunction with ChamQ Universal SYBR qPCR master mix (Vazyme Biotech, Nanjing, China). The 16S rRNA gene served as an internal control, and relative gene expression levels were quantified utilizing the 2^–∆∆CT threshold cycle (CT) method. All qRT-PCR primers are listed in Table 2. All samples were run in triplicate.

2.11. Homology modeling

To further evaluate the possible effects of specific amino acid mutations on protein structure, 3D structural homology models of AsDddY was generated using Swiss-Model (http:// wissmodel.expasy.org/). Molecular docking was visualized by AutoDock Vina software (Li et al., 2017), the figure was generated by PyMOL software.

2.12. Accession number(s)

The draft genome sequence data of strain ZS25 has been deposited at NCBI under the accession number JBMBWU000000000.

3. Results

3.1. Isolation and identification of DMSP-degrading strain

Using 5 mM DMSP as the sole carbon source, DMSP-catabolizing bacteria were isolated from the sediment of Zhoukou Park. Approximately 10 strains that could grow on MSM plates containing 5 mM DMSP were isolated. Among these bacterial strains, Strain ZS25 grew well in the medium containing DMSP or acrylate as the sole carbon source (Fig. S2). Furthermore, gas chromatography (GC) and HPLC analysis showed that ZS25 could catabolize DMSP to produce DMS and acrylate, respectively (Fig. 2A and Fig. S3). Based on the 16S rRNA gene sequences (Fig. 2B), strain ZS25 exhibited high similarity to the genus Acinetobacter, particularly with Acinetobacter baylyi DSM 14961^T, showing 99.86 % similarity, confirming strain ZS25 as Acinetobacter sp. ZS25.

Fig. 2.

The utilization of DMSP by Acinetobacter sp. ZS25. (A) GC detection of DMS production from DMSP by strain ZS25. The culture medium without bacteria was used as the control. The DMS standard was used as a positive control. Acinetobacter sp. ZS25 could catabolize DMSP and produce DMS; (B) Neighbor-joining tree based on 16S rRNA gene sequences showing the phylogenetic relationship between ZS25 and the most closely related species. Accession numbers of sequences are given in parentheses. Bar, 0.005 substitutions per nucleotide position.

3.2. Degradation pathway of DMSP in strain ZS25

Strain ZS25 transformed DMSP into DMS, undetected in the substrate control without added strain ZS25. To elucidate the genetic components implicated in the degradation of DMSP within Acinetobacter sp. ZS25, we conducted genomic sequencing and performed a comparative analysis to identify homologs of established DMSP lyases. A comparative genomic analysis between strain ZS25 and other known DMSP-degrading strains, revealed the presence of several key genes in the chromosome of strain ZS25 that were associated with the degradation of DMSP. Specifically, the dddY gene, which encoded dimethylsulfoniopropionate lyase that catalyzes the initial DMSP to DMS and acrylate, the acuNK genes responsible for converting acrylate to 3-hydroxypropionate (3HP), the dddA and dddC gene, encoding iron-containing alcohol dehydrogenase and CoA-acylating methylmalonate-semialdehyde dehydrogenase, respectively, responsible for metabolized 3HP into malonate semi-aldehyde (Mal-SA) and acetyl-CoA, and a TetR/AcrR family transcriptional regulator named acuR, were identified. DddY, AcuN, AcuK, DddA, DddC and AcuR showed highly similarities of 81 %, 85 %, 84 %, 34 %, 57 %, and 39 %, respectively, with the counterparts in the well-investigated DMSP-degrading strains, Alcaligenes faecalis M3A (Fig. 3A).

Fig. 3.

The function of Acinetobacter sp. ZS25 dddY in DMSP metabolism. (A) Genetic organization of the putative DMSP-catabolizing gene cluster in strain Acinetobacter sp. ZS25 and Alcaligenes faecalis J481; (B) Quantitative transcriptional analysis of dddY, acuI, dddC, dddA, acuK and acuN in strain ZS25 in the presence of 5.0 mM glucose or 1.0 mM DMSP. The transcriptional level of the 16S rRNA gene was used as an internal standard, and the data in each column were calculated by the 2^–∆∆CT method using three replicates. (C) Detection of DMS production from DMSP degradation by the wild-type strain ZS25, the ZS25ΔdddY mutant, the complemented mutant, the complemented mutant ZS25ΔdddY/dddY, and the mutant complimented with an empty vector pBBR1MCS-5.

3.3. Functional verification of dddY

To determine whether DMSP has an inducible effect on the dddY-acu cluster of strain ZS25, dddY, acuI, dddC, dddA, acuK and acuN was selected, and investigated by reverse transcription-quantitative PCR (RT-qPCR). The mRNA levels were compared in strain ZS25 grown on glucose or DMSP. In strain ZS25, the transcription levels of dddY, acuI, dddC, dddA, acuK and acuN increased by 12.2-fold, 9.2-fold, 8.3-fold, 9.1-fold, 6.6-fold, and 7.1-fold, respectively, in cells grown on DMSP compared to glucose (Fig. 3B).

To further investigate the role of the dddY gene in strain ZS25, a mutant strain with the dddY gene deleted was constructed and designated as ZS25ΔdddY. The experimental results indicated that the ZS25ΔdddY mutant strain exhibited a partial loss of the ability to convert DMSP to DMS (Fig. 3C), suggesting the presence of an alternative DMSP biodegradation pathway in strain ZS25. Subsequently, a complementary strain, designated as ZS25ΔdddY/dddY, was developed. The complemented strain ZS25ΔdddY/dddY restored the ability to convert DMSP to DMS. These findings support the hypothesis that dddY encodes an enzyme with in vivo DMSP lyase activity, consistent with the observed phenotype following the knockout of the dddY gene in strain Alcaligenes faecalis M3A.

3.4. Expression and characterization of dddY

The dddY gene of strain ZS25 encodes a polypeptide consisting of 1209 amino acids. To investigate the in vitro DMSP lyase activity of ZS25 DddY, the full-length dddY gene was amplified and cloned into the pET29a(+) plasmid for expression in E. coli BL21 (DE3) cells. Subsequently, the recombinant DddY protein was purified. SDS-PAGE analysis confirmed the production of the anticipated 45.9-kDa protein, designated as AsDddY, following IPTG induction, thereby demonstrating the successful expression and purification of AsDddY (Fig. 4A).

Fig. 4.

Characterization of recombinant AsDddY. (A) SDS-PAGE of purified DddY stained with Coomassie Brilliant Blue. Lane M contains protein standards; lane 1, purified C-terminal His-tagged DddY protein; (B) HPLC detection of acrylate production from DMSP lysis catalyzed by the recombinant AsDddY. The reaction system without DddY was used as the control. (C) Effect of temperature on the enzymatic activity of AsDddY. The activity of AsDddY at 50°C was defined as 100 %. (D) Effect of pH on the enzymatic activity of AsDddY. The activity of AsDddY at pH 8.0 was defined as 100 %. (E) Non-linear fit curve for DMSP cleavage by AsDddY. The kinetic parameters were determined under pH 8.0 at 50°C.

GC and HPLC analysis showed that pure AsDddY catalyzed the release of DMS and acrylate, respectively, from DMSP, confirming that AsDddY was a functional DMSP lyase in vitro (Fig. 4B). Regardless of pH and temperature conditions, the enzyme system fails to produce acrylate in the absence of AsDddY, thereby confirming that DMS production is dependent on enzyme activity (Fig. S4A and Fig. S4B). The optimal temperature for AsDddY enzyme activity was 50°C (Fig. 4C), which was slightly lower than the 60°C previously reported for A. bereziniae DddY (AbDddY, sharing 86.85 % sequence identity) and higher than the 37–40°C previously reported for A. faecalis DddY (AfDddY, sharing 81.89 % sequence identity), while values for these DddY enzymes was the same at pH ∼8.0 (Fig. 4D). This phenomenon may be intricately linked to the environmental conditions under which strain ZS25 proliferates. Strain ZS25 was isolated from the sediment of Zhoukou Park. The elevated temperatures characteristic of the city, coupled with the humid and oppressive conditions of the sedimentary environment, likely contribute to the distinct properties of AsDddY. However, AsDddY maintained 48.4 % of its highest enzymatic activity at 30°C and 88.4 % at 40°C (Fig. 4C), and AsDddY maintained 91.4 % of its highest enzymatic activity at pH ∼7.0 (Fig. 4D), indicating that AsDddY is a viable enzyme in physiological environments.

The recombinant AsDddY exhibited a K_m value of 2.6 mM for DMSP at pH 8.0 and 50°C (Fig. 4E), which is lower than that of the reported DMSP lyases AbDddY (5.06 mM), but higher than that of AfDddY (1.4 mM) and DaDddY (0.4 mM) (DddY from strain Desulfovibrio acrylicus). The observed relatively high K_m values, which fall within the millimolar range, are characteristic of various DMSP catabolic enzymes, including the DMSP lyases DddK, DddP, DddQ, DddU, DddW, and Alma1 (Alcolombri et al., 2015; Brummett et al., 2015; Curson et al., 2018; Hehemann et al., 2014; Li et al., 2014; Sun et al., 2016; Todd et al., 2011; Wang et al., 2015), a as well as the DMSP demethylase DmdA (Reisch et al., 2008b). These values substantially surpass the nanomolar concentrations of DMSP typically present in seawater, indicating that DMSP may play essential physiological roles within bacterial cells. Then the k_cat value of AsDddY was tested, approximately 12.7 × 10³ s⁻¹ (Fig. 4E), which was higher than that of AbDddY (8.0 × 10³ s⁻¹), AfDddY (0.3 × 10³ s⁻¹) and DaDddY (2.07 × 10³ s⁻¹), suggests that strong conversion capacity of AsDddY in DMSP metabolism.

3.5. Key amino acid residues mutation of dddY

Previous research has elucidated that the mature form of DddY is structurally characterized by two distinct domains: the N-terminal domain, which extends from residues Ala22 to Val190, and the C-terminal domain, spanning residues Ser191 to Pro401. The N-terminal domain is predominantly composed of α-helices and envelops the C-terminal domain in a manner akin to a "cap," thereby being referred to as the "cap domain" (Li et al., 2017; Wang et al., 2015). Conversely, the C-terminal domain primarily consists of eight antiparallel β-strands, forming a β-barrel fold structure that accommodates the catalytic pocket of DddY, thus termed the "catalytic domain." Bacterial DMSP lyases are categorized into several distinct superfamilies, with DddK, DddL, DddQ, DddY, and DddW sharing a conserved cupin motif, thereby classifying them within the cupin superfamily (Fig. S5). This study examines the cupin superfamily, which comprises a functionally diverse group of proteins that generally require a divalent metal co-factor for their enzymatic activity. Using inductively coupled plasma atomic emission spectrometry, it was determined that approximately 68 % of DddY molecules are associated with Zn²⁺ ions (Andrew R J Andrew R J Curson et al., 2011; Li et al., 2017). For a comparative analysis of amino acid sequences, we selected a range of representative proteins from the cupin superfamily. Specifically, twelve conserved amino acids-Thr131, Asp181, Tyr225, Gly230, Gly250, His263, His265, Glu269, Tyr271, Leu274, Tyr331, and His338—were mutated to alanine to evaluate their impact on protein activity. The results demonstrated that mutations at His263, His265, Glu269, Tyr271, and His338 resulted in a near-complete loss of activity in the DddY protein compared to the wild-type strains. In contrast, mutations at Thr131, Asp181, Tyr225, Gly230, Gly250, Leu274, and Tyr331 retained partial activity, with respective activity levels of 20.6 %, 31.2 %, 7.5 %, 28.9 %, 24.7 %, 14.6 %, and 11.4 %. These findings indicate that amino acids located within the two cupin motifs are particularly critical for maintaining the functional integrity of DddY (Fig. 5A and 5B).

Fig. 5.

Structural and mutational analyses of AsDddY. (A) Enzymatic activities of the mutants of AsDddY at pH 8.0 and 50 °C. The activity of wide-type DddY was defined as 100 %; (B) Structural analyses of important residues in the active site of AsDddY, the cupin motif 1 was shown in yellow and cupin motif 2 was shown in purple. The mutant residues were shown in sticks.

Additionally, the enzymatic activities of the twelve AsDddY mutants were systematically assessed across a range of temperatures and pH levels. All mutants exhibited maximal activity at a pH of 8.0 and a temperature of 50°C. Under optimal pH conditions (pH 8.0), the mutants were inactive at 70°C and demonstrated diminished activity at 10°C, 20°C, 30°C, and 60°C, while retaining relatively high activity at 40°C. Conversely, under optimal temperature conditions (50°C), the mutants were inactive at pH levels of 4.0, 5.0, and 10.0, showed reduced activity at pH 6.0 and pH 9.0, and exhibited relatively high activity at pH 7.0. These characteristics are consistent with those observed in the wild-type AsDddY (Fig. S6 and S7).

3.6. Analysis of dddY-acu gene clusters of A. sp. ZS25 and related bacteria

The dddY-acu cluster responsible for DMSP degradation was initially identified in the A. faecalis M3A strain (genes D6I95_14,815 to D6I95_148,850), with the pathway and the function of each gene now well elucidated. Subsequently, homologous genes have been identified in the A. bereziniae and Desulfovibrio acrylicus strains (Li et al., 2017; Maarel et al., 1996), as well as ZS25 in this study. The configurations of the dddY-acu cluster in these four representative strains exhibit considerable variation. To explore the diversity of the dddY-acu cluster in other bacterial species, DddY was employed as a query in public databases, NCBI, to identify potential strains with available genome sequences containing the dddY-acu cluster. Subsequently, the arrangements of all identified dddY-acu clusters were systematically analyzed. A detailed database survey showed that the dddY gene, is widely present in bacteria. However, not all strains containing the dddY gene have a complete dddY-acu cluster. The number of strains with a complete dddY-acu cluster is limited. The findings indicated the identification of 19 distinct gene order types, including A. sp. ZS25 (Genbank accession number SAMN47294383), A. Bereziniae PUMA0285 (SAMN41379045), A. faecalis J481 (SAMN10101723), A. Sacchari HC-19 (SAMD00797420), V. sp. tm (SAMN46144076), D. Halophila SFB-3 (SAMN11950944), Azoarcus sp. PHD (SAMN11950943), A. Pittii A9028 (SAMN20080703), A. sp. WK018 (SAMN38635465), S. benthica (SAMN24537884), G. apicola ESL0182 (SAMN08297256), R. trehalosifermentans H1998003092 (SAMN05892269), X. sp. KK7.4 (SAMEA115015404), A. calcoaceticus TUM15316 (SAMD00175054), P. sp. P5 E6 (SAMN42323802), Pseudomonadota bacterium D5M12_12 (SAMN47248198), Zoogloeaceae bacterium HK-STAS-PROT-111 (SAMN15687996), Aestuariibacter sp. A3R04 (SAMN19350919). Based on the gene count, the configuration of the dddY-acu cluster in these representative strains can be categorized into four distinct types (Fig. 6A). Notably, strain A. sp. A3R04 contains two homologous DddY genes, and further investigation is required to determine whether both genes possess the ability to cleave DMSP. In strain A. pittii A9028, the homologous genes acuN and dddY are separated by a substantial distance, approximately 33,000 base pairs. Similarly, in strain A. bereziniae PUMA0285, the complete dddY-acu cluster is partitioned into two segments, with the dddAC gene located at a considerable distance from the dddY gene, approximately 50,000 base pairs apart.

Fig. 6.

Co-evolutionary analysis. (A) The arrangements of acu-dddY gene cluster in strain ZS25 and 18 other bacteria; (B) Phylogenetic analysis based on 16S rRNA gene sequences; (C) Phylogenetic analysis based on DddY amino acid sequences.

Then based on the 16S rRNA gene sequences and DddY protein sequences, the phylogenetic trees of the 19 strains mentioned above were constructed (Fig. 6B and 6C). It was found that the evolutionary relationships of the 19 strains based on the 16S rRNA sequences were very similar to those based on the DddY proteins. These two phylogenetic trees were mainly divided into three parts. The upper part (shaded in light pink color) contained 4 kinds of bacteria which almost belonged to genus Acinetobacter, while the middle and lower part (shaded in light blue and green color, respectively) contained five and six kinds of bacteria, respectively, almost belonging to belonged to beta-proteobacteria. These findings suggest that DddY proteins co-evolved with their hosts, but not horizontally.

4. Discussion

The degradation mechanism of the dddY-acu cluster involved in DMSP catabolism has been extensively studied over the years. This research trajectory began with Yoch et al.'s 1995 discovery of the DMSP-degrading strain Alcaligenes faecalis J481, which harbors the dddY-acu gene cluster, and continued until 2011, when the mechanism of action of DddY was first elucidated. In this study, strain ZS25 identified as Acinetobacter species was a DMSP-degrading strain that contained dddY-acu cluster as well, increasing the variety of DMSP-degrading strains in the environment. The enzymatic properties of DddY were systematically characterized, including the determination of optimal temperature and pH conditions, as well as the calculation of the K_m and k_cat values. Furthermore, the critical amino acids of the AsDddY protein were identified. The research also examined 19 configurations of dddY-acu clusters, highlighting the diversity inherent in these clusters. AsDddY represents the fourth homologous protein of DddY to be characterized, thereby enhancing the comprehension of the DMSP degradation mechanism mediated by DddY proteins.

In this study, the mutation of the dddY gene in strain ZS25 resulted in only a partial loss of its capacity to convert DMSP into DMS and acrylate. However, the ability to degrade DMSP was restored upon reintroduction of the dddY gene into the mutant strain ZS25ΔdddY. These findings indicate the presence of one or more unidentified DMSP degradation pathways in strain ZS25. The existence of multiple DMSP degradation pathways within a single strain is further corroborated by observations in other strains. For instance, in strain J481, mutation of the dddY gene similarly led to only a partial reduction in its ability to transform DMSP, as seen in the mutant strain J481ΔdddY (Andrew R J Andrew R J Curson et al., 2011). In the strain Amylibacter cionae H-12, mutations in the dddU gene resulted in the partial loss of the ability to degrade DMSP. Conversely, certain strains possess only a single gene cluster responsible for DMSP degradation (Wang et al., 2023b). For instance, in the strain Psychrobacter sp. D2, mutation of the dddX gene led to a complete loss of DMSP degradation capability (Li et al., 2021). The presence of multiple degradation pathways facilitates the survival of these strains in DMSP-rich environments, exemplifying environmental adaptation. However, to date, no additional DMSP degradation genes have been cloned in strains J481, H-12, or D2, highlighting an urgent need for further investigation.

The presence of the acuI gene in strain ZS25 is noteworthy. The acuI gene encodes an MDR family oxidoreductase. Our investigation into 19 diverse dddY-acu gene clusters revealed that 16 of these clusters contained the acuI gene in all configurations, whereas the dddY gene was present in only three configurations. This indicates that in the majority of strains possessing dddY-acu clusters, both dddY and acuI genes are present, suggesting their involvement in the upstream degradation pathway of dimethylsulfoniopropionate (DMSP). Furthermore, quantitative real-time PCR (qRT-PCR) experiments demonstrated that DMSP induces the expression of the acuI gene. To further elucidate the role of the acuI gene in strain ZS25, we generated an acuI gene knockout mutant, designated ZS25-dacuI, and assessed its ability to transform DMSP. The findings indicated that mutations in the acuI gene diminished the capacity of strain ZS25 to convert DMSP, as illustrated in Figure S8. It was demonstrated that the dddY and acuI genes function synergistically within strain ZS25. These results are consistent with observations in strain J481, where mutations in the acuI gene similarly impaired the strain's ability to degrade DMSP. However, a notable distinction is that in strain J481, the acuI gene is adjacent to the dddY gene, forming a single operon, whereas in strain ZS25, the dddY and acuI genes are located distantly and oriented in opposite directions. Further investigation is required to elucidate the left-right mechanism of AcuI within the dddY-acu gene clusters.

Among all DMSP lyases, DddY is uniquely situated in the periplasmic space. In this study, the optimal temperature for AsDddY was determined to be 50°C, with an optimal pH of 8.0. In our investigation of the enzymatic properties of AsDddY, we employed two distinct buffer systems: Tris–HCl buffer, with an effective pH range of 7.0–9.0, and Britton-Robinson buffer, which covers a broader pH range of 1.8–11.9. For determining the optimal pH of AsDddY, we utilized Britton-Robinson buffer to adjust the pH range from 4.0 to 10.0, due to the limited buffering capacity of Tris–HCl buffer. It is important to acknowledge that the discontinuity between these two buffer systems may introduce variability in the experimental results. However, our findings indicated that at pH 8.0, there was no significant difference in the efficiency of AsDddY-mediated conversion of DMSP to acrylate when either Tris–HCl buffer or Britton-Robinson buffer was used to adjust the pH. Consequently, within this study, the discontinuity of the buffer system may introduce certain inaccuracies; however, these are likely to be minimal. The K_m value was measured at 2.6 mM, which is relatively high compared to many DMSP catabolic enzymes. This high K_m value notably exceeds the nanomolar concentrations of DMSPs typically present in seawater, suggesting that DMSPs may play crucial physiological roles within bacterial cells. The k_cat value for AsDddY was approximately 12.7 × 10³ s⁻¹. For comparison, the k_cat values for other characterized DddY enzymes were as follows: AbDddY at 8.0 × 10³ s⁻¹, AfDddY at 0.3 × 10³ s⁻¹, and DaDddY at 2.07 × 10³ s⁻¹. It is noteworthy that all DddY enzymes identified to date exhibit significantly higher k_cat values compared to other DMSP lyases. The k_cat value of AsDddY was 18,142-fold, 42,333-fold, 12,700-fold, 694-fold, 14,111-fold, and 1814-fold higher than those of DddU, DddP, DddQ, DddW, DddK and Alma1, respectively (Alcolombri et al., 2015; Brummett et al., 2015; Kirkwood et al., 2010; Li et al., 2017; Peng et al., 2019; Wang et al., 2023b). The periplasmic localization, which is in close proximity to environmental DMSP, along with the high k_cat values observed for DddY, strongly suggests that this enzyme plays a crucial role in DMSP metabolism, facilitating the production of acrylate.

Among the DMSP lyases, DddK, DddL, DddQ, DddW, and DddY are classified within a cupin-like superfamily. Despite variations in amino acid composition and protein structure among these five types of Ddd proteins, they all share two cupin motifs. The first motif is characterized by four fully conserved residues, HxHxxxxExY (His263, His265, Glu269, and Tyr271; numbering corresponds to AsDddY), while the second motif contains an additional fully conserved histidine, His338. In this study, mutations in DddY involving these five amino acids resulted in a near-complete loss of its ability to transform DMSP. These findings are consistent with previous observations of AbDddY in strain A. Bereziniae and PlDddQ in strain R. Lacuscaerulensis ITI_1157 (Li et al., 2014, 2017). Prior research has identified three amino acids His265, Glu269, and His338 as being integral to Zn²⁺ coordination. Nowadays, some scholars have pointed out that the conserved Tyr of motif 1 (Tyr271, numbering is for AsDddY) may not be essential for DMSP lyase activity. The Y277A mutant of DaDddY retains about 20 % of wild-type activity, indicating that its contribution to catalysis is minor. Zhang et al. suggested that this tyrosine is retained in DMSP lyases such as DddY or DddL, but its role in catalysis is relatively small, and it seems more likely that this active site tyrosine plays a key role in other functions or functions of cupin-dll (Li et al., 2017). In this paper, the DddY protein retains 2.1 % of its activity after mutating Tyr271, which further confirms this hypothesis.

Research has demonstrated that the DddY enzyme in E. coli is capable of cleaving both DMSP and DMSOP. Further investigations have revealed that, in addition to DddY, five other DddY lyases—namely, DddL, DddQ, DddW, DddK and DddU—also exhibit DMSOP lyase activity, even in the presence of equimolar concentrations of DMSP (Carrión et al., 2023; Chhalodiaa and Dickschat, 2023). This observation is not unexpected, given the structural similarities between DMSP and DMSOP, as well as the presence of a cupin domain and a similar catalytic mechanism in these DMSP lyases. Notably, among these lyases, only DddY and DddL demonstrate higher in vivo activity against DMSP than DMSOP when present at molar concentrations equivalent to the two substrates (Carrión et al., 2023; Chhalodi and Dickschat, 2024). Additionally, it is noteworthy that beyond the six aforementioned DMSP lyases, DMSP hydrolases from various bacteria, fungi, and algae—such as type III CoA transferases DddD, acyl-CoA synthetase DddX, M24 metallopeptidase DddP, and the aspartate racemase Alma1 superfamily DMSP lyases—also exhibit DMSOP hydrolase activity. Purified DddP, Alma1, and cupin domain Ddd enzymes degrade DMSOP into DMSO and acrylate. Furthermore, DddX is capable of converting DMSOP into DMSO, although the by-product, presumably acryloyl-CoA, has not been consistently identified. In particular, when the proteins DddP, Alma1, DddY, and DddX are incubated with equimolar concentrations of DMSOP and DMSP, they generate between 1.4 and 5 times more DMS than DMSO (Burkhardt et al., 2017; Chhalodi and Dickschat, 2024; Li et al., 2017). This observation indicates that the proteins DddY, DddL, DddP, DddX, DddY, and Alma1 exhibit greater catalytic activity against DMSP compared to DMSOP, warranting further investigation into the underlying mechanisms of this phenomenon. Subsequently, research by Jeroen S. Dickschat et al. demonstrated that DMSP lyases not only process their native substrates, DMSP and DMSOP, but also facilitate the cleavage of various analogs of these compounds. Nevertheless, DMSOP and DMSP remain the most efficacious substrates for the transformation of their respective analogs. Additionally, the enzymes DddW, DddK, and DddY exhibit greater efficiency in converting DMSP analogs than their corresponding DMSOP analogs (Carrión et al., 2023; Chhalodi and Dickschat, 2024). Beyond the aforementioned six DMSP lyases, there is a lack of studies investigating the activity of the DddD, DddX, DddP, and Alma superfamily against DMSOP analogs, highlighting a need for further exploration in this area.

5. Conclusion

In this study, Acinetobacter sp. ZS25 capable of mineralizing DMSP was isolated and identified. Metabolite analysis revealed a dddY-acu metabolic pathway for DMSP degradation in strain ZS25. The functions of dimethylsulfoniopropionate lyase DddY was identified through comparative genomics, heterologous expression, and gene knockout experiments and the key amino acid residues were investigated. Moreover, 19 distinct acu-dddY cluster order types were identified via Bioinformatics analysis. This research provided new insights into the microbial degradation mechanism of DMSP via dddY-acu cluster and have a better understanding of diversity of dddY-acu clusters.

Funding information

This work was supported by the National Natural Science Foundation of China (No. 32400106), the Projects of the Joint Fund of Henan Province's Science and Technology Research and Development Program (Industrial Category) (No. 225101610054), the Department of Science and Technology Planning Project of Henan Province (No.252102110328) and the Natural Science Foundation of Henan Province (No.252300421695).

CRediT authorship statement

Yongchuang Liu: Writing-rewiew and editing, data analysis, writing original draft, Conceptualization, funding acquisition and supervision;

Feilong Ma: Writing-review and editing;

Hongfei Zhang: Writing-review and editing;

Cuiwei Chu: Conceptualization;

Xiaohui Wang: Writing-rewiew and editing, data analysis;

Yuehui Tang: Writing-rewiew and editing, data analysis;

Jian He: Writing-review and editing;

Jiguo Qiu: Writing-review and editing;

Siqiong Xu: Writing-review and editing; funding acquisition and supervision;

Lili Li: Writing-review and editing; funding acquisition and supervision;

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability

No data was used for the research described in the article.